Google is rolling out a device-level RCS handshake verification system on Android to combat AI-powered voice deepfake scam calls by cryptographically confirming caller identity before the phone rings.
Aviva is using AI-driven detection systems to stop £230M in sophisticated insurance fraud, including emerging threats from deepfake imagery and synthetic claim evidence.
New research shows synthetic benchmarks overstate the scaling potential of Forward-Forward and layer-local training methods, revealing real-data limits that challenge claims about backpropagation alternatives.
A new position paper argues that a true science of AI requires studying training dynamics directly rather than relying on post-hoc fixes, with implications for understanding generative models and synthetic media systems.
AI-generated influencers and synthetic content creators are flooding social platforms, blurring the line between authentic and artificial personalities and challenging viewers' ability to distinguish real from fake.
A school shooting survivor is suing an AI gun detection company after its computer vision system failed to flag a visible weapon, raising critical questions about AI detection reliability, vendor accountability, and the limits of automated visual recognition in safety-critical deployments.
Harness-1 is a 20B-parameter retrieval subagent trained with reinforcement learning inside a stateful search harness built on gpt-oss-20b, designed to handle multi-step search and retrieval tasks more reliably than prompt-only agents.
Bolster AI is presenting its deepfake defense and unified brand protection platform at the Gartner Security & Risk Management Summit, signaling growing enterprise demand for integrated synthetic media detection tools.
Reality Defender is scaling its enterprise deepfake detection footprint through an AWS Marketplace listing, a ZeroFox integration for social media threat intelligence, and recognition in a new Gartner report on synthetic media risk.
Andrej Karpathy compresses 33 years of neural network evolution into a 90-second retrospective, tracing the architectural lineage from early MLPs to today's transformer-based generative models powering synthetic media.
A hands-on exploration of training GPT-style transformer models on a budget GTX 1050 GPU, revealing practical constraints, optimization tricks, and what hobbyists can realistically achieve with limited VRAM.
New research suggests LLM safety alignment degrades significantly as prompt length grows, exposing a structural weakness in how guardrails generalize beyond short, training-distribution inputs.